BACKGROUND
Empathy is a difficult thing to foster in children. There are a range of theoretical methodologies that can be applied to how children learn to be empathetic. Some would say “My child is naturally empathetic.” Others would say that empathy is learned through socialisation at pre-school.
Regardless of how it is developed, it is a vital ingredient in the child’s toolbox, allowing them to cope with and discover strategies that will forever assist them in their connections with other people and their own personal growth.
This project was inspired by a conversation with a friend. She said:
“Aw! I’m feeling very inspired up here, although these boys get an amazing life, I need to figure out how to show them contrast, ngahere grew up with some poor years, these guys have no clue, it’s hard to teach how lucky they are”
NNvB, sms message. 21 January 2020
Originally, I developed a narrative-driven approach to tell 3 stories where the child-user would be put in the shoes of another, trying to understand the decisions, reactions and choices another child may make. Get the kid to see that there is a lot happening behind any choice or decision any individual makes. And to be able to ‘read’ that choice or decision from ‘their shoes.’ This was complex…overly so. Once I began wireframing it, it resembled a discarded plate of spaghetti, and was near impossible for me to create without the use of a game engine.
To me, wealth was an important part of the conversation. This was reinforced when researching this project. Writing in Psychology Today, Sam Osherson states
“[The] wealthy have more difficulty reading other people’s emotions and difficulty understanding the fears, hopes, and problems of people different from them.”
https://www.psychologytoday.com/intl/blog/listen/201410/how-teachers-can-reduce-the-empathy-gap-wealth-creates
Gaming and game theory also became an essential part of this process. I wanted to create a game where kids between 8-12 could literally ‘walk in the shoes’ of another child. This fast became too complex for the time period (though would make an intriguing longer-term project), so I had to reduce. After some quick surveying, I discovered that parents believed children develop empathy far earlier than I has assumed, at around 4-5. I refined the story to appeal to younger kids, and a much simpler storyline.
This was also overly complex, so I returned to one of the origins of the idea, a friends ‘Feelings Wheel’ she had on her fridge that helped her kids describe their emotions. I wanted to put this feeling wheel into action.
RESEARCH
The background research for this project carried on throughout the short timeframe from ideation to delivery. Each day I learned something new, which complicated not only my thinking around the project, but also what I could, or felt like I could achieve.
One of the first discoveries was the browser-based game SPENT, which focused on the choices people are forced to make when living in poverty. I believed this model was what I wanted to build my game on.
The academic material (reading list at the bottom of the page) I engaged with was complicated, but basically reinforced the same message, if you want to design something for kids, involve them in it! In terms of simplifying the game, Johan Waldemarsson puts it simply and concisely:
“keep it relatively simple and accessible but also don’t be afraid to challenge the child and make the game too easy” (p.4)
http://ltu.diva-portal.org/smash/get/diva2:1016184/FULLTEXT02.pdf
Waldermarsson’s focus was a more medical approach, but the impact of his work on mine made it simpler. Reducing the cognitive load for both the audience of the game, and myself simplified my approach to the product.
“carefully see if you can implement relevant information and possible quizzes where it’s appropriate, without removing the fun of the gaming experience.” (ibid)
Understanding the development of empathy in kids from a psychological perspective was central to this project. While there was no central ‘rule’ in terms of a childs’ empathetic development, there are frameworks available:
- Empathy is first developed by mimicking emotions 0 to 2.
- 3 to 4 kids are aware of their feelings and some are aware of the impact on others
- 5 to 6. Kids learn to show compassion and a greater application of empathetic values.
(see https://www.psychologytoday.com/intl/blog/smart-parenting-smarter-kids/201905/how-children-develop-empathy and https://www.scholastic.com/teachers/articles/teaching-content/ages-stages-empathy/)
SURVEY
I developed a survey through Google Forms that ran from 2nd February – 19th Feb 2020, which yielded 45 responses. I kept the survey running while I was in the development phase as I was interested in the ongoing verbatim responses. This provided great insights into what people thought, and forced me to reconsider both the premise of my project, and the approach I was taking. The survey was sent out to friends, and run through a variety of child-focused groups on Facebook, including rainbow family pages.
“It would be great to see a book for 5 yr olds, it’s an age that they seem to really struggle with the idea of empathy.”
Survey Respondent
43 of the 45 were from New Zealand, with one each from Australia and the USA. One person wasn’t a parent, but interested in the project.
The purpose of the survey was 2-fold. Firstly, to see if there was interest in a game-based tool that could develop empathy for children, and secondly, what such a tool would look like.
My assumption at the time was that wealth was a influencing factor in a child’s empathetic development. This assumption was developed by my own peer group, who are often complaining “My kids don’t know how lucky they have it.”
Based on these sort of conversations with friends and family, I created a list of the common threads that arose. The 9 ways to teach empathy were:
- Games
- Sharing stories
- Books
- Friends/peer group
- Visiting other people’s homes
- Being responsible for a pet
- Being responsible for a sibling or friend
- Putting another person first
- Seeing things through someone else’s eyes
93.2% of participants said that developing empathy was important or very important for children’s development, and of the 9 ways of possibly teaching empathy, 94% of respondents said being able to see things through someone else’s eyes was the most important and it could be achieved mostly by sharing stories (94%). 81.8% of respondents were also interested in participating in feedback sessions and/or testing the prototype.
This confirmed both my approach, and the goal I had with the product.
In terms of what terms could be applied to understanding empathy, ‘Understanding’ was the most important term, with all parents indicating it, followed by ‘compassion’ and ‘feelings’. ‘Social status’ (13.6%), ‘Money’ (9.1%) and ‘Wealth’ (11.4%) ranked lowest, which is a contrast to the academic view that empathy is harder to develop for wealthy kids.
The majority of parents (59.1%) believed empathy was developed in children between 0 and 4, meaning I needed to angle my product to a younger market.
The roadblocks the parents surveyed found when teaching children could be broken down into 7 main areas:
- The child’s ego: “Him not understanding others perspective” “Inherent selfishness of kids.”
- Short attention spans: “It is VERY hard for children to imagine anything other than their own immediate experience or comprehend that someone else feels differently.”
- Parent’s behaviour (time): “I try to model empathy to them but their dad seems to have lost all empathy in his life which isn’t helping”
- Not having the right words (kids): “Not knowing how to explain it properly, and missing teachable moments”, “Using the right language to help them understand”
- Peer pressure (kids): “Our boys have short attention spans so things need to be explained quickly. And when en masse, their bravado, being “too cool” to listen.”
- Social conditions: “Societal norms, especially when considering a childs individual sense of identity, being their own person.”
- Lack of resources: “Most books, games, and videos do not take an empathetic stance so it’s hard to find media that aligns with our views.”, ““As a Pakeha, finding resources (books etc) about other people’s situations in life that don’t seem to have an underlying message / tone that white people are better.”
CODESIGN WORKSHOP
After changing the age focus I was going for, I contacted 2 parent-friends to test the 2 narratives I had developed. One was a mother of 2 boys (10 and 7), the other was a grandmother to 2 girls (10 and 5). While this wasn’t an ideal sample size, the insights were strikingly similar. My personal relationship with the children could have also influenced their responses.
- All children liked the younger story and could relate to it easier. The 10-year-old boy said that the older story made him sad. The 10-year-old girl said the older story “sounded like some of her friends.” She was referring to the kid living in his Aunt’s garage with his Dad.
- They liked the pick-a-path approach, but wanted more “video”.
- The younger children asked their parent/caregiver which path they should take most the time. The older children paused, and look for affirmation before pressing their choice.
- Three of the children said they “liked” the aesthetic, but I feel they were aware I was testing something I had made. One (10-year-old boy) said he thought it was boring.
- When asked how they would make it, the younger kids said “Robots” and “Puppies” would make it better.
- Both adults wanted more teachable moments.
Changes after codesign workshop
After engaging with the kids and their parents, I realised there was some problems with the language. Kids understand ‘You’ better than when ‘you’ are being someone else (as in the first narrative).
I added the robot and puppy option to the second story, and simplified it to a reductive version of what it was. This changed the way each character’s story flowed. It also allowed a broader approach to teachable moments, with the addition of colour theory, simple math and spelling complimenting the empathy-teaching. I also added a lot of moments where the child would be instructed to “talk with an adult you trust”.
These changes were implemented in the narrative, but the prototype followed only one day of one story.
USER TESTING
After considering all the changes needed, the time available, and the goal of the project (to teach empathy in kids) the scope of the project changed to be more focused on teaching young kids to be able to identify emotions (as a precursor to understanding ‘what another person is feeling’) and a far more simplified prototype was developed.
The revised prototype was tested with 2 children under 6. One remotely (where I watched their interaction through Zoom) and one face-to-face (the 5-year-old from the codesign workshop). Accessing more children in the time available wasn’t possible due to my own commitments, but more the multiple commitments of the targeted group. Kids are very busy!
Child 1, 6, understood the tasks in front of him, and checked his answered with his dad before clicking his option. He said he liked the icons (drawn from emoji’s), because they looked “like Daddy’s messages”. He could sound out the words on the buttons, but needed his father’s help to understand what the tasks expected of him were. Therefore, the text panels are too long!
Child 2, aged 5, wasn’t as keen on the look of the game. She said, “The other one is better.” To me, this reinforced the more complicated visual direction of the second prototype. She was confused by some of the emotions represented, such as the distinction between angry and mad, and asked why a lot. Her grandmother did a fantastic job of explaining the more complex emotions to her, giving her examples from the last week (she had just started school) to describe the differences.
Both children commented on the black-and-white approach, as they didn’t know what was able to be clicked to advance in the game. Both children also found identifying similar feelings hard. For example, there is an emotional difference between being angry upset, and sad upset. While the use of colour signified this (blue for sad and red for mad), combining this with closely-related emotions was harder to identify.
This informed the final iteration, where the use of levels and sticking to a similar structure to the feelings wheel was employed to give a sense of development in the game.
Design Experiments
This project has had a range of design experimentation along the way. Each iteration was a reduced version of the one before that, and changed the project almost entirely.
Version 1: Too Complex
The first iteration of the story was very involved and complex. You can read the narrative structure here.
The characters were developed in conversation with a teacher, who identified different sorts of students she had experienced in her classes. I reduced the different characters into 3 archetypes and focused on what their story would look like. Each character was given a background, personality and had different drivers. My hope with this was that the older children the game was originally aimed at (7-12) would pick one character they related most to.
I sent the story outlines to 3 parents, and it tested well. One said she found the story of the boy living with his Dad in the garage tough, but knew of people in that situation, and had to explain it to her child once.
The reward structure in the game was in three parts.
- Money: players could make and spend money. The goal was to have money left at the end of the month.
- Feelings: each choice or interaction impacted on the players ‘feelings’ bar.
- Social standing: each choice or interaction impacted on the players ‘social’ bar.
With the complicated scoring structure, it meant that each interaction could have 4-5 different outcomes. In a game engine, this would have been a simpler fix, but the limitations of prototyping tools meant I couldn’t just insert an equation.
Really rough wireframe 
Kumera Springs sketch 
Characters
I devised a look-and-feel, sketching out all the characters, and the town. These were simplified characters that I intended to vectorise and colour later. Originally I wanted to do the game in watercolour, but this quickly added a level of complexity I didn’t need.
Narrative wireframing was done in Figma, as this proved to be the most useful way to ‘see’ the scale of the game. It was vast. 71 screens and only a week into the game. This wasn’t going to be possible.
“It is VERY hard for children to imagine anything other than their own immediate experience or comprehend that someone else feels differently.”
Survey Respondent
I was also impacted by some of the responses coming in from the survey. It was obvious I had aimed the story for older children than parents thought was appropriate for children’s empathetic development.
The first iteration was abandoned, and I restructured the narrative with the codesign responses in mind.
Version 2: Simple, but can’t be done in time
Simplify the story in terms of narrative and time was suggested in class, and at first I was resistant to it, but realised to achieve any sort of delivery, it needed to be reduced.
One of the biggest changes was to change the language to YOU from being in someone else’s position. The revised story is here:
Key points from the revised story:
- ‘Look after’ someone/thing, not ‘be’ it.
- Time is one week, not one month.
- Removed scoring all together. The reward is in the emotions represented on the thing you are looking after’s face (allowing empathetic/feeling recognition).
- Make it more for younger kids.
- Include ‘teachable’ moments (this was colour theory, spelling and math)
- Reinforce the need to talk to an adult you trust if you’re unsure
While in the story all three characters weeks and events were outlined, in the prototype I followed one line of the story, Prince the Puppy. Visual resources were taken from free sites which removed the added pressure of designing every visual (bad move on reflection).
Even this approach was going to be too complicated in the timeframe and with the competing pressures. I went through one day in the time with Prince, which resulted in 25 screens, 14 for the intro, and 11 for the day.
Version 3: So simple, and a return to the Feelings Wheel
After the class with Celia Hodent, I was truly inspired. She talked a lot about cognitive development, the importance of Jost Law (distributed learning) and the importance of learning by doing. She discussed how reducing the cognitive load was central to a decent learning experience and that there would always be trade-off in any game development. She said how important it was to not tie rewards to money (phew, in my case) and that punishing behaviours wasn’t ideal. There was so much learning in my head that day I was overwhelmed and inspired at the same time. It made me step back and critically revise my project.
I has made so many assumptions in the development of this product, and needed to return to perhaps the simplest option: the feeling wheel I had seen in my friend’s kitchen.
Iteration 3, simply called the EmoSpotter, is a heavy reduction from the original concept, but purely focused on the assisting a child’s ability to identify the complex web of emotions through expression recognition.
In the paper Facial Expression Recognition: A Survey (Yunxin HuangOrcID,Fei Chen,Shaohe Lv andXiaodong Wang, 2019. https://www.mdpi.com/2073-8994/11/10/1189/htm) the authors discuss how there are basic emotions, complex emotions and micro expressions. These factors can be applied to the development of emoji’s as a simplified way to express complex emotions in a digital space.
In Emoji as a Proxy of Emotional Communication (Guillermo Santamaría-Bonfil and Orlando Grabiel Toledano López, 2019. https://www.intechopen.com/online-first/emoji-as-a-proxy-of-emotional-communication) the authors talk about how the simplified language of emoji’s have become a pan-national mode of expression. I wanted to take this thinking and apply it to the feelings wheel, giving kids a way to feel a little adult as some kids recognised the emoji’s from text messaging while identifying what that emoji represented.
Based on a simplified feelings wheel, I designed standardised faces that represented 21 emotions. Colour was used to signify if it was Mad, Sad or Happy and there was a tree of emotions that dived further into the complexity of what that emotion could mean.
The first interface was simple, too simple, and after 2 user tests, needed to be better designed. The learning from user testing is mentioned above.
The second iteration of EmoSpotter provides a level-up approach to understanding what the emoji’s and words mean. It is more colourful than the first, and has the addition of complex emotions, or the limitations of the English language; how Upset can be a mad sort of upset, or a sad sort of upset. The only person who has tested it said that as a tool, it could also be used to help people on the spectrum to identify what their emotions are. This was flattering and something I never considered.
The full end-to-end prototype is here.
Results & Conclusions
Learning
One of the biggest things I learnt over the course of this project was that I really, really enjoy researching. Trying to ask the right questions to not only answer whatever problem statement is before you, but also being able to analyse the information you’re provided with is central to the UX process. This applied in both this paper, and the Advanced UX project (which was heavily research-focused).
I would have preferred to engaged with the participants of the survey to refine the product before developing it. I could have spent a month focusing on what sort of product parents wanted alone!
Limitations
Time has been the biggest limitation. Having just over 3 weeks to deliver this project, from concept inception to research to design and development and the iterations needed along the way has been the biggest limitation. I am over time, and know I will be penalised for that. Perhaps in future, a 50% project could be released earlier and given the time it deserves? I would like to have been able to had a sort of narrative arch in my own product development, seeing the impacts of one thing impact the other.
Another limitation has been of my own making: refining my ideas. The original story would have taken months in development with nothing else happening around. I should have been aware of that.
Trying to apply academic research, which often runs years behind technology has been another limitation, but this was balanced by the class with Celia Hodent. That learning is still peculating in my mind.
Conclusion
Despite the almost constant iterations and reductions that have happened across this project, I’m more aware of what I really enjoy in terms of UX. I want to do things that matter, that can impact some sort of social change, and believe I have angled my choices of projects to meet this end.
I hope to be able to spend more time in a thought space…discovering and defining opportunities for applying UX growth to real-world issues.
There is always something that has to give, and I’m again reminded of the first day of class…Done is better than perfect.
Jeremiah Boniface 